Thinking out loud about Microsoft making Windows an open source project is a great way to get your friends and colleagues wondering seriously about your mental health. It's an idea strange enough to sound practically paradoxical, like "hot ice" or "short Pink Floyd songs."
It's intriguing for the same reason, however and people both inside and outside of Microsoft have begun to openly consider the potential upsides and downsides of such a move.
+ ALSO ON NETWORK WORLD: Cross Microsoft off the list of potential Salesforce suitors -- and maybe everyone | Patch Tuesday may be dead, but Microsoft's not confessing to the crime +
But it's not a big one, by most estimates. Network World contributor and SUSE employee Bryan Lunduke's estimation that Microsoft releasing Windows' assets and code via something like the GPL "would, in all likelihood, happen at the exact same moment that all volcanoes on Earth turn into waffle cones filled with chocolate ice cream" seems to be where the smart money currently is.
Network World contributor and SUSE employee Bryan Lunduke
"The fact that this has even the slightest chance of ever happening boggles the mind," he said.
Why in the world would Microsoft do this?
Broadly speaking, to gain all the traditional advantages of open source software community developed code that has more eyes out for bugs, goodwill from developers, and potential user base growth, thanks to favorable ($0) pricing on the base product.
Michael LaVista is the founder and CEO of Caxy, a Chicago-based web and UX design firm that uses open source software. He said that a newly opened Windows code base would have its advantages.
"One of the huge benefits of the open source community approach is that issues are spotted quickly and patched quickly," he said. "The downside is seeing what's in there. This is old news, but some source code was famously leaked and it had some unfortunate comments in it. Who knows how much is still in there[?]"
Others were also at pains to highlight the double-edged nature of going open source. Jono Bacon, who used to be the community manager for Ubuntu, said that open source projects differ even among themselves in their balance of efficiency and openness.
"Open source provides a wonderful opportunity to broaden your engineering base, but there is a cost in terms of development efficiency," he said.
Bacon held up Debian as an example of a project heavily focused on openness and collaboration everyone has a voice, everyone can contribute, but democracy can slow down decision-making. By contrast, he said, Canonical administered Ubuntu rather differently still open, but with the company making most of the major design and development choices.
"If Microsoft were to build Windows as an Open Source project, I suspect they would take the Canonical approach rather than the Debian approach," he told Network World.
In essence, the open source Windows would be a Linux distribution, developed by a community but guided from above by Microsoft.
And if that guidance is sufficiently hands-off, the company could win itself a lot of brownie points among developers. While the days of the company's near-total hegemony over personal computing are long past, many still remember that time less than fondly.
An open source Windows would be an even bigger deal than the recent opening of the .Net framework, argues LaVista.
"Right now I'd argue most startups won't use .Net for a variety of reasons including cost, but also that it's corporate,'" he said. "Open source changes the model from top-down to community. That's a huge shift but a huge opportunity."
That appeal to startups and smaller businesses could prove more fruitful than a renewed focus on the enterprise market, LaVista added.
"I think the enterprise market is largely stuck," he said. "There would be a huge switching cost."
Why in the world wouldn't Microsoft do this?
Obviously, reasons No. 1 and 1a are money and uncertainty. Despite rapidly declining revenue figures for Windows, Microsoft still makes billions of dollars from OEM and personal license sales.
And even though not all of that money would vanish if Windows were to become available for free, the transition to a services and support revenue model could be a difficult one.
That said, the transition could be made, according to Bill Weinberg, senior director of open source strategy for Black Duck Software, a software and consulting firm in Burlington, Mass.
"[H]owever Redmond would choose to monetize a FOSS Windows, it has additional revenue sources in its adjunct product lines, most notably [Microsoft] Office and other applications," he told Network World.
More to the point, however, it's the fact that Microsoft simply isn't prepared to operate as a benevolent backer to an open source version of Windows that makes the possibility so remote. Weinberg highlighted that the task of building the infrastructure to support a project of such scale is massive.
"It took Linux a decade to build its behemoth development and support network," he said. "And the success of the Linux OS can be attributed to a unique mix of developer enthusiasm, timely corporate sponsorship, choice of license and good will."
Jim Zemlin, executive director of the Linux Foundation, underlined the point with authority, saying that the range of infrastructure required might surprise even Microsoft with its extent.
An open source Windows, he said, would need a formalized governance structure, a legal framework for handling patents and certification programs and the like, a social coding infrastructure, and code structured for for collaborative development for starters, anyway.
"We have no doubt the developers who write Windows are talented but how does the code look under the covers?" Zemlin asked. "Is it architected in a way that is modular to enable mass collaboration?"
OK, so where does that leave us?
Pretty much right where we started from. The upshot is that there are some intriguing potential upsides for Microsoft in open-sourcing Windows developers would love it, the user base could get broader, revenues are generally headed south anyway, so why not throw the doors open and possibly even get a better-designed piece of software into the bargain?
But there are at least as many downsides, which make the prospect of it happening anytime soon wildly unlikely the scope of the undertaking, from an organizational and coding standpoint, would be massive even by Microsoft's standards. The benefits might not be worth the extraordinary amount of work.